Revised sampling designs by each group

Write in your revised sampling design in the table below

As well as acting on the comments on the designs above (look at other group's work for comments), and discussions in class,
also take care to add what results you would expect
  • under the null hypothesis (drift, gene flow, weak/no selection)
  • under the alternative hypothesis (drift, gene flow, some selection)


Sampling Method:
We will have one horizontal line transect with three sample areas (each at least 20m apart) in shrub lands and grass. On arrival to the field we will choose a horizontal strip abundant in both scrubland and grassland, we will chose this area when we find at least one large populations of snails and conduct the line transect from there . We have combined with another group who will do a horizontal line transect at another altitude.

Advantages:
  • Each sampling area is 20m apart so the data will not be affecting by gene flow
  • Choosing to sample at one altitude eliminates altitude as a variable
  • Combining with another group who will carry out a line transect at a different height provides us with more data and helps us to see if altitude is an important factor in determining snail habitat and selection processes
  • We have specifically chosen sampling areas instead of random sampling so that we can see clear results in selection between grassy and shrub areas and to ensure each sample area consists of large populations that we can compare.
  • We have chosen an equal number of sampling areas for both grass and shrub and no woodland so that we have as much data as possible to compare between the two areas.
  • By choosing three sample areas in both the shrub and grass areas we are able to replicate our results in order to increase its reliability.

Disadvantages:
  • There may not be a horizontal line of 3 areas of both grass and shrub land
  • Not being completely random could bias our results in an unforeseen way
  • Choosing a single line may not be a true representation of the whole area that we are sampling in.
  • By collaborating with another group the human error in the results may be greater.



Contributing members:
Rupert Higgins
Gemma Shireby
Christina Pope
Jacqui Moneke
Jessica Patricot